Separate blitz ratings?

Running Chess Tournaments
1

Will there now be a separate, third rating system for Blitz (G/3 to G/9)? Or has this been decided yet?

I certainly hope so. The argument that not enough Blitz games are submitted for rating to establish a reasonably large database does not impress me. I say, get the thing going anyway, and let it grow gradually.

This brings up another question. If G/30 to G/60 is dual-rated Regular and Quick, will G/10 to, say, G/15 be dual-rated Quick and Blitz?

Again, I hope so. For one thing, this would partially defuse the not-enough-games argument.

Bill Smythe

2

The Delegates did not consider the issue of Blitz ratings. The Executive Board has the authority to make such a decision, their next meeting is in late September. I think they’re currently leaning towards keeping just two OTB ratings systems but encouraging Blitz events.

The Delegates did authorize a replacement for the current ‘additional famly member’ category.

There will be two Family Membership plans:

Plan 1: $64, covers both spouses, all minor children living in the household (ie, at the same address), plus any full-time college students under age 25 (even if living away from home.)

Plan 2: $35, covers all minor children living in the household (ie, at the same address).

A Family Membership includes 12 issues of Chess Life, including the six scholastic inserts that members age 10 and under began receiving in August.

The office is authorized to accept the new family memberships as soon as the membership programming is updated to handle it, probably by September 1st.

3

In time blitz will be the third rating. Even the federation understands that blitz and quick are two different systems. After the end of the WBCA, the federation did not have any plans to have a third rating. Hope some time in 2005 that the federation will have blitz as a rating.

If the federation does have a blitz rating, that it makes the time of the blitz be between G/3 and G/9. That it makes it clear what the time take back: if the director takes take away from a blitz clock for a 2 second delay.

As most chess clubs do have during the meetings have a blitz tournament, with a entry fee of only a few dollars; the rating fee for these blitz tournaments should be much lower or the clubs will not bother sending the tournament report. As the smaller chess clubs that are affilated with the federation, would only have a certified club tournament director and would be more or less sending the tournament report on paper – the cost of 40 cents per game would have to be placed onto the players. If a player that plays in a blitz tournament and sent in on paper, and the player plays 10 games would be a total cost of one [10 games (40 cents / 2 players) = $2] player be $2. Each person that plays a game, the cost to rate the event is 20 cents as needing two players to make one game. With the club needing to make a little money for a blitz event, the need to have the event rated would in the long run force chess clubs to ask for more money for USCF Blitz tournaments. The rating fee of a blitz event has to be low or the clubs will never perform one.

The second problem with a USCF Blitz tournament, that most blitz tournaments do not use the Crenshaw-Berger tables when it comes down to color. Most blitz tournaments are round robin, but the players at the board choice in what matter they choice for color not the tournament director. That a blitz event only would have better luck if it is a double round robin, then it would not matter when one player did get white on the first game or the second – during the tournament every player had equal amounts of white and black. If it is a one round Crenshaw-Berger without the director making the claim of color, and let the players choice the colors: then someone could have all white during the tournament or all black during the tournament.

The other factor with the federation can only go up to 20 rounds, as the computer program can not rate any event with more then 20 rounds. If a blitz tournament has 22 players in a Crenshaw-Berger table, then the tournament would have 21 rounds and the federation could not rate the event. It is just a software problem, the federation needs to address this problem with modern software before the federation can make any claim of having a blitz rated events.

Earnest,
Douglas M. Forsythe

4

Well, now that you’ve decided the future of the USCF, I guess that will save the USCF Executive Board the time to consider the merits and costs of supporting a third independent OTB rating system. :slight_smile:

Also, I don’t know what blitz tournaments YOU play in, but every one I’ve ever directed, played in or seen has either used the proper tables for color (or some other non-arbitrary system for assigning color) or was a double round robin.

And since when did the word ‘choice’ become a verb?

5

Well even if for some reason run for the USCF Executive Board, think would still get more votes then Sam Sloan. It might not be this board but some board in some time will have a third independent OTB rating.

Earnest,
Douglas M. Forsythe

6

hmm, lets see if I can conjugate this verb: To Choice
I choice
You choice
He/She/It choices

We choice
You(y’all) choice
They choice

I choice to send LDTF back to grammar school!! I also choice to demand that he upgrade to Senior, if only for the puproses of sending him back to Club level!!!

7

ok, I apologize for the personal attacks against you, Douglas. It’s not my style, but damn it’s so difficult reading your posts, man!! Please get a grammar checker!!

8
9

Well, now that English 101 is over for the day, :slight_smile:, returning to the base subject, I don’t know that a sufficient case has been made for either the necessity or viability of a third independent OTB rating system for blitz chess.

Yes, we should be able to do it fairly soon under the new programming, but the question is why should we?

Not all things that are possible are good ideas, or necessary.

Walter Browne’s blitz ratings didn’t survive in the real world, why would the USCF succeed where he failed?

Mike Nolan

10

USCF is the standard for American Chess.

Blitz isn’t going to be as popular as maybe Quick or Regular, but what would the overall cost be to implement something like that?

What percentage of tournaments are Quick versus Regular? Would players rather play in a nightly tournament with 4 rounds or 10?

I wish I would have been able to go down to Florida this past week and figure out what’s going on with USCF and all the financial stuff.

11

The major reason that it failed as people had to join the WBCA. Chess Clubs will have USCF members that play over-the-board chess. The only problem with having a USCF blitz tournament is the rating fee. As most club blitz events are only a few dollars, nothing more then five dollars. If the club also has to make sure the event would be USCF rated they would have to add in the rating fee, then make sure the director or directors become certified tournament directors.

Earnest,
Douglas M. Forsythe

12

I posted some statistics on regular and quick rated events the other day. As I recall, less than half of the quick-rated events (about 1600 sections in calendar 2003, maybe 10% of the total number of sections rated) were quick-rated only, the rest were dual-rated.

I have my suspiciion that if the original intent of the Delegates (that all games in the G/30 to G/60 window be dual rated) was strictly followed, that the percentage of quick-only events would be much lower.

There are many factors that would enter into determining the cost of a separate blitz rating system.

First there’s the time needed for deciding what kind of system it should be. The regular and quick ratings systems are not identical, though they are very similar. Should a blitz system use the same formulas as one of those systems or some other set of formulas?

I’m sure the Ratings Committee would design such a system if requested to do so. I don’t know how long they would take to complete such a task.

A separate issue, one mentioned by another poster, has to do with which events would be rated under which system. I don’t know that this is exclusively a Ratings Committee decision. Similarly, what rules should apply to USCF-rated blitz events? What membership requirements would apply?

Then there’s the cost of changing the programming to accomodate a blitz system. This is (mostly) a one-time cost to change the forms to accept another rating system choice and to write and test a computational module that conforms to the standard developed by the Ratings Committee.

MSA would have to be changed to show blitz ratings, which means that several programs would need minor modifications.

As far as I can tell, SwisSys and WinTD can accomodate the USCF’s two rating systems, but would have to be updated to handle independent blitz ratings.

The ongoing administrative cost is another issue. How do blitz ratings get published? Do we have a separate supplement, a separate section in the supplement, no supplement (e.g., only available online), or do we add a third column to the existing supplement? What about the ratings supplement files?

Having blitz ratings would probably increase the number of events that get rated under the wrong system, thus it would increase the number of events that have to be fixed. The good news here is that the new programming is designed to make this a much more automated process than it has been in the past.

Having some idea of the one-time and ongoing costs of the system doesn’t really address the viability issue. Would this reduce the number of quick games being rated (already a small fraction of the number of regular-rated games). Would TD’s run many USCF-rated blitz events?

So I ask again: Walter Browne’s blitz ratings failed in the marketplace, what would lead us to believe that the USCF can succeed where he failed?

13

This is an extremely legitimate point. With the present rating fees, a 16-player round robin would cost 3 times as much as a 16-player 5-round Swiss. Perhaps a new fee structure would be in order. Instead of 40 cents per game, it could be 20 cents per game plus 50 cents per player, for example. For that matter, a structure like this could be used for Regular and Quick, as well as Blitz.

In my experience, most blitz tournanents DO use the Crenshaw-Berger tables for colors, even when they don’t use them for round numbers. The Crenshaw-Berger color rule is extremely simple: Both odd or both even, higher number has white. Odd vs even, lower number has white. Almost everybody uses this, even when the games are played in random order.

This can be changed. Mike Nolan has probably already increased the limit to 40 games or so, in his new software.

Bill Smythe

14

It wasn’t the blitz ratings that failed, it was WBCA itself. With USCF, an existing body would simply be adding a third rating system.

Bill Smythe

15

What are the differences? The only ones I can think of are (1) a player’s regular rating is used to initialize his quick rating, and (2) there may be a different K-factor. The issues generated by such differences should not take long to resolve.

Again, a very brief discussion among two or three committees should answer this question. An obvious start would be to dual-rate (quick and blitz) events from G/10 through G/15 (or thereabouts).

These are all worry-wart type concerns. Of course, effort would be involved, but none of this seems insurmountable.

The fact that USCF already has ratings, and would not require membership in a second organization in order to give a person a blitz rating.

Bill Smythe

16

This is a red herring. The ED has already declared that WBCA-like rules (such as illegal move loses, and no delay) can be used by organizers of blitz events as long as these rules are announced prior to round 1.

Which set of rules is used for blitz has nothing to do with whether there are combined quick-blitz ratings or separate ratings.

Bill Smythe

17

In the blitz tournaments that for one reason or the other have been with, when dealing with the Crenshaw_Berger tables has been the first person that enters the event are ranked number one, second enters ranked number two till all the players are in the event. As it is a club tournament they do not use the USCF ratings for ranking. If it does become a USCF event then if it is a one round event then they would need to use the USCF rating for rankings.

As the players are at a chess board, or between games finding the next person to play. They know they have to play each other, they could go to the crosstables and check there ranking on the Chrenshaw-Berger: then check the correct color in the tables in the official rule book. Or they can be at the table, one person picks up the white pawn and black pawn, then mixed the pawns around behind their back, then let the other player pick the hand that would be the color.

If it is a USCF rated blitz tournament, the club would need a certified tournament director (most small clubs do not have any members that are certified as tournament directors); they would have to have the rankings based on USCF ratings not the order of players that pay the rating fee; that the colors must be assigned by the tournament director, not a judgement call of the players.

The other problem for the clubs would be some people that do play in blitz tournaments are not or are they willing to joint the USCF. Would force the clubs in having a all USCF members tournament and non-USCF members tournament. If the club has to pass on the rating fee onto the USCF members tournament, would they not also be forced to pass on the cost of the rating fee to the non-USCF tournament. If not then the USCF blitz tournament would be a $6 entry fee and the non-USCF tournament could be the old cost of $2. If this happens only the die hards would be in the USCF tournament, everyone else in the non-USCF tournaments. If it is only a USCF blitz tournament, some clubs could lose 30 to 80 percent of the regulars. Then the club would lose that entry fee of say $2 each and every week that would have helped get needed equipment.

Earnest
Douglas M. Forsythe

18

And that’s the best way to do it, whether or not the event is USCF-rated. It randomizes the colors better. Otherwise, the club’s top-rated player will always have white against his strongest opponent, week after week.

There is NO requirement that, if a round-robin is USCF-rated, ratings be used to determine wall chart order.

WRONG. See above.

Or, they could just notice, at the start of the tournament, which number they are. The conversation typically proceeds along the following lines: “Are you in the tournament?” “I sure am. I’m number 5.” “I’m number 8, so you have the white pieces. Let’s do it!”

A sign like the following is typically posted at blitz tournaments: “Odd-vs-odd or even-vs-even, larger number has white. Odd-vs-even, smaller number has white.” This saves a lot of confusion and running to the pairing tables.

BAD, BAD, BAD. A player could end up with many more whites than blacks, or vice versa.

Bill Smythe

19

There is not a good system or bad system with a round robin tournament. If we look at a round robin or Crenshaw-Berger with 4 players as most are common with a quad. Round 1, 1-4 2-3 round 2, 3-1 4-2 round 3 1-2 3-4, the director on the third round can determin colors by a toss: most directors just keep with the same crosstables.

Why (Bill Symthe) should a quad need to have rankings by rating when a blitz should not – if and only if they are both USCF rated? If you say that the top player would have white against the second tope player that would be true. This is the reason that a blitz USCF event has to be a double round robin. If some tournament switch the colors around on the Crenshaw-Berger, then all the players one week will always have to play without adding more players the next. Even if all twenty players play one week does not mean they will all play the next week, without having someone wanting to play the next week.

Not that many people are going to look at table K on page 306 of the 5th edition if there are 24 players. Each club has there own rules and have there own ideas how to choice colors. The only logical way to speed up a blitz tournament, is having a double round robin. Then the tournament director after the tournament can check the ratings and format the rankings, then place for section 1 win, draw, lost then section 2 win, draw, lost. The players only have to tell the director on game one the score and game two the score. If it is a one section blitz event, then colors will always be a problem.

Earnest,
Douglas M. Forsythe

20

Of course they’re worry-wart details, but the details are what makes a system work or not. A few points:

  1. There is no inherent upper limit in the number of rounds in the new data structures. From a practical consideration, I don’t know that anything larger than 32 makes much sense.

  2. From the information processing end of things, there’s no justification for lowered fees for blitz games versus non-blitz games. They should all cost roughly the same amount of time to upload, validate and rate. There could be differing administrative support issues, just as there could be different rules for tournament play.

Those may not significantly affect the programming of the system but they may well affect the overall cost of administering an additional system, and that could affect the decision as to whether to support an independent blitz system as well as whether or not to dual rate some events as both blitz and quick.

  1. I think you correctly identified all of the differences between the quick formulas and the regular formulas, but ANY differences whatsoever will be dealt with by using separate computational subroutines. In actuality there are already three ratings systems in the new system: Regular, Quick and Correspondence. Though we wouldn’t have a computational module for them, we could add another row in the table for a FIDE-rated player to store his or her FIDE rating.

  2. The discussions in Fort Lauderdale on Regular vs Quick ratings mostly came down to one issue: Are ‘regular’ ratings better than ‘quick’ ratings?

Given that there are those who think quick ratings should be adjusted periodically to keep them in sync with regular ratings and that more than a few TD’s indicated that they use the regular rating for pairing purposes even if an event is only going to be quick-rated, that would seem to indicate that the answer to the question is ‘yes’. As a result, I’m not sure that the quick rating system is viable over the long term either.